Avamar

A System Administrator’s Guide to Monitoring

A System Administrator’s Guide to Monitoring

Avamar 

Avamar 

Executive Summary

Executive Summary

Backup and recovery infrastructure is critical to the life of any business. Longer data Backup and recovery infrastructure is critical to the life of any business. Longer data retention requirements and legal regulations for data storage and retrieval contribute to retention requirements and legal regulations for data storage and retrieval contribute to the need for a robust but agile backup solution. Given strict recover time and recover the need for a robust but agile backup solution. Given strict recover time and recover point objectives that define many service level agreements, the backup team operates point objectives that define many service level agreements, the backup team operates around the clock. This group must be conscious of costs and often work with

around the clock. This group must be conscious of costs and often work with

constrained budgets. A monitoring solution needs to be dependable and cost effective. constrained budgets. A monitoring solution needs to be dependable and cost effective. EMC Avamar 

EMC Avamar ®® is a software/hardware backup and recovery solution that takes is a software/hardware backup and recovery solution that takes advantage of data commonality in the environment and deduplicates the data at the advantage of data commonality in the environment and deduplicates the data at the source client and globally at the Avamar server. Avamar detects changes at the subfile source client and globally at the Avamar server. Avamar detects changes at the subfile level and uses local cache files to speed file comparison and processing. The result is level and uses local cache files to speed file comparison and processing. The result is fast client-side processing, low network bandwidth utilization (as only the changed fast client-side processing, low network bandwidth utilization (as only the changed blocks are sent to the Avamar server), and lower storage consumption. Although blocks are sent to the Avamar server), and lower storage consumption. Although  Avamar includes mechanisms for self-alerting, external monitoring should be added.  Avamar includes mechanisms for self-alerting, external monitoring should be added.

Nagios is a powerful enterprise-class monitoring system that enables organizations to Nagios is a powerful enterprise-class monitoring system that enables organizations to quickly identify and resolve infrastructure problems. This Open Source tool

quickly identify and resolve infrastructure problems. This Open Source tool——with awith a large community of developers and support

large community of developers and support——is easily customizable for your situationis easily customizable for your situation and add-ons can be created to meet your needs.

and add-ons can be created to meet your needs.

Syslog-NG is a leading Open Source solution for log collection and management. It Syslog-NG is a leading Open Source solution for log collection and management. It provides a central logging set up that can be used to filter messages with tools such as provides a central logging set up that can be used to filter messages with tools such as SWATCH.

SWATCH.

Introduction

Introduction

This article will describe a general solution for monitoring the Avamar environment from This article will describe a general solution for monitoring the Avamar environment from a Systems Administrator viewpoint using Nagios and Syslog. It will highlight some of a Systems Administrator viewpoint using Nagios and Syslog. It will highlight some of  Avamar’s

 Avamar’s builtbuilt-in utilities and focus on how they can interact with external tools.-in utilities and focus on how they can interact with external tools.

Audience

Audience

This article is intended for Avamar administrators with knowledge of UNIX/Linux This article is intended for Avamar administrators with knowledge of UNIX/Linux administration. It is not intended as a step-by-step guide for installing and configuring administration. It is not intended as a step-by-step guide for installing and configuring each package but to make the user aware of the capabilities.

each package but to make the user aware of the capabilities.

Avamar Architecture

Avamar Architecture

 Avamar is a disk-based backup solution. The Avamar software installs on top of  Avamar is a disk-based backup solution. The Avamar software installs on top of provided hardware or qualified customer hardware. This article will focus on the provided hardware or qualified customer hardware. This article will focus on the

EMC-provided data storage solution. Hardware configurations can consist of single node or provided data storage solution. Hardware configurations can consist of single node or multiple node implementations. A node is a physical server running the Red Hat Linux multiple node implementations. A node is a physical server running the Red Hat Linux operating system. The two basic nodes are the Utility node and the Data node. The operating system. The two basic nodes are the Utility node and the Data node. The Utility node provides internal Avamar server processes and services, including the Utility node provides internal Avamar server processes and services, including the administrator server, jobs scheduling, authentication, and maintenance and web administrator server, jobs scheduling, authentication, and maintenance and web

access. The Data node is dedicated to providing storage for the backups. In an Avamar access. The Data node is dedicated to providing storage for the backups. In an Avamar single node implementation, the Utility and Data node functions reside on the same single node implementation, the Utility and Data node functions reside on the same physical node. In a multiple node configuration, the Utility node is its own dedicated physical node. In a multiple node configuration, the Utility node is its own dedicated physical node. The amount of capacity and protection needed will determine the physical node. The amount of capacity and protection needed will determine the number of Data nodes.

number of Data nodes.

Avamar Notification Features

Avamar Notification Features

 Avamar system activity and operational status is reported as various event codes to the  Avamar system activity and operational status is reported as various event codes to the Utility node MCS service. Examples of events include maintenance activity status, failed Utility node MCS service. Examples of events include maintenance activity status, failed logins, or client activations. Each event contains useful information such as the event logins, or client activations. Each event contains useful information such as the event code, date and time stamp, category, type, summary, and what generated the event. code, date and time stamp, category, type, summary, and what generated the event. Notification options can consist of email messages, pop-up alerts, Syslog, SNMP, and Notification options can consist of email messages, pop-up alerts, Syslog, SNMP, and events sent directly to EMC support. For this article, we will use the Syslog fe

events sent directly to EMC support. For this article, we will use the Syslog fe ature toature to achieve our desired results. Notifications are configured as Profiles via the Avamar achieve our desired results. Notifications are configured as Profiles via the Avamar Console.

Console.

Setting up the SYSLOG environment

Setting up the SYSLOG environment

The building blocks for setting up the SYSLOG monitoring environment will be in this The building blocks for setting up the SYSLOG monitoring environment will be in this order:

order: 1.

1. Identifying Identifying a server to a server to act as act as the Monitoring the Monitoring serverserver 2.

2. Setting Setting up up the the SYSLOG-NG SYSLOG-NG serverserver 3.

3. Setting up Setting up SWATCH to SWATCH to filter the filter the SYSLOG-NG messagesSYSLOG-NG messages 4.

4. Configuring Configuring SYSLOG SYSLOG on ton the Avamar he Avamar nodesnodes 5.

5. Setting Setting up tup the Avamar he Avamar SYSLOG ProfilSYSLOG Profilee

You will need a new or existing Linux/UNIX server in your environment that can be used You will need a new or existing Linux/UNIX server in your environment that can be used as the central monitoring server. This sever can be a virtual machine or physical server. as the central monitoring server. This sever can be a virtual machine or physical server. You need to install the base operating system such as Red Hat, Solaris, HPUX, or

You need to install the base operating system such as Red Hat, Solaris, HPUX, or flavor of your choice. For details on installing the OS and packages, please consult the flavor of your choice. For details on installing the OS and packages, please consult the respective OS manuals.

respective OS manuals.

Download a copy of SYSLOG-NG for your OS to your central monitoring server. You Download a copy of SYSLOG-NG for your OS to your central monitoring server. You will need to disable the syslog service that comes with your OS before installing. Install will need to disable the syslog service that comes with your OS before installing. Install the SYSLOG-NG package. You can edit the syslog-ng.conf file to customize it with you the SYSLOG-NG package. You can edit the syslog-ng.conf file to customize it with you preferences. For example, you can have SYSLOG-NG create a directory for each host preferences. For example, you can have SYSLOG-NG create a directory for each host that sends it a message and name the log file using your standard naming convention. that sends it a message and name the log file using your standard naming convention. Here is an example:

Start the SYSLOG-NG process.

Start the SYSLOG-NG process. Verify the process is running by using the ―ps‖Verify the process is running by using the ―ps‖ command.

command.

SWATCH, a free tool that uses pattern matching to filter log messages, will monitor log SWATCH, a free tool that uses pattern matching to filter log messages, will monitor log files as they are written in real-time. As patterns are matched, SWATCH will then carry files as they are written in real-time. As patterns are matched, SWATCH will then carry out specific actions as defined in the configuration file. Those actions can be simple out specific actions as defined in the configuration file. Those actions can be simple email notifications or execution of a script. Download a copy SWATCH for your OS email notifications or execution of a script. Download a copy SWATCH for your OS platform and install as per the platform.

platform and install as per the platform.

Now we must edit the syslog-ng.conf file to have it ca

Now we must edit the syslog-ng.conf file to have it call SWATCH when a new messagell SWATCH when a new message arrives and process it. Here is an example:

arrives and process it. Here is an example:

 After the syslog-ng.conf file has been updated, we will need to config

 After the syslog-ng.conf file has been updated, we will need to configure the SWATCHure the SWATCH configuration file to look for certain patterns and send notifications. You can use regular configuration file to look for certain patterns and send notifications. You can use regular expressions for pattern matching. Please see a sample Avamar SWATCH configuration expressions for pattern matching. Please see a sample Avamar SWATCH configuration in the Appendix. In this example, we are watching for the Avamar code ―22605‖, which in the Appendix. In this example, we are watching for the Avamar code ―22605‖, which is a Node Offline event. We use the keyword

is a Node Offline event. We use the keyword watch for watch for ..

We then can use the keyword

We then can use the keyword mail mail  or or pipe pipe to act upon that match. In this example, we to act upon that match. In this example, we are using the

are using the pipe pipe keyword to send the text of the message to a homegrown internal keyword to send the text of the message to a homegrown internal script called

script called notify notify which then sends it along to a recipient.which then sends it along to a recipient.

The complete stanza would look like this: The complete stanza would look like this:

Here is an example of the EMAIL alert message: Here is an example of the EMAIL alert message:

The SYSLOG-NG service needs to be restarted each time that the swatchrc file is The SYSLOG-NG service needs to be restarted each time that the swatchrc file is

updated. SWATCH does have the ability to point the swatchrc configuration file against updated. SWATCH does have the ability to point the swatchrc configuration file against an existing log file for testing.

an existing log file for testing.

 Avamar now needs to be configured to take advantage of the central monitoring server.  Avamar now needs to be configured to take advantage of the central monitoring server.

We must add a line to the /etc/syslog.conf file on

We must add a line to the /etc/syslog.conf file on ALLALL the Avamar nodes (including the the Avamar nodes (including the spare node on multiple node implementation). This allows us to capture all system spare node on multiple node implementation). This allows us to capture all system messages regardless of if the Avamar application is running. It also provides a central messages regardless of if the Avamar application is running. It also provides a central location for archiving syslog messages. The entry in the syslog.conf file on each

location for archiving syslog messages. The entry in the syslog.conf file on each

 Avamar node needs to point to the central monitoring server. For example, if the central  Avamar node needs to point to the central monitoring server. For example, if the central

monitoring server had IP 1.2.3.4, it would look like this: monitoring server had IP 1.2.3.4, it would look like this:

You will then need to restart the syslog service on the Avamar node using the command You will then need to restart the syslog service on the Avamar node using the command ―service syslog restart‖.

―service syslog restart‖.

Back on the central monitoring server, in the location you configured for the host Back on the central monitoring server, in the location you configured for the host directories to be created, you should now see a directory for this Avamar node. That directories to be created, you should now see a directory for this Avamar node. That directory will contain a log file with a new entry.

directory will contain a log file with a new entry.

To capture Avamar application-generated syslog events, we need to create a profile in To capture Avamar application-generated syslog events, we need to create a profile in the Avamar console. Launch the Avamar console. From the Tools menu, click Manage the Avamar console. Launch the Avamar console. From the Tools menu, click Manage Profiles.

Profiles.

Click on Local Syslog to highlight and then click Copy. Save the new profile. Click on Local Syslog to highlight and then click Copy. Save the new profile.

Highlight the new profile and click Edit from the top menu. Click on the Syslog Notification tab.

Enter the IP address of the central monitoring server. The port will stay at the default of 514. Check the ―Include extended event data‖ checkbox. This option causes the alert message to be delimited using tags for easier parsing by SWATCH. Click OK.

Highlight the new profile you created and then click Enable from the top menu.

 A syslog central monitoring system is now in place. The benefits to this setup include:

  ALL default Avamar events are sent to central monitor for processing. This

eliminates the need to pick and choose what alerts you want Avamar to send from the long list.

 In a multiple Avamar grid environment, you wouldn’t have to log in to each

console and modify events that are no longer needed as this is done centrally.

 Syslog on the physical nodes will send server-based events outside of the

 Avamar application that might be specific to that hardware platform.

 In a multiple node configuration, the Spare node is now monitored using Syslog

so that you can fix any issues before the need arises to use it.

  A central SWATCH configuration provides a quick way to add alerts or

notification options.

Nagios Overview

Nagios is a graphical monitoring application that can easily provide tactical overviews of your environment. You can monitor physical devices as well as process and services running on those devices. You can further customize Nagios by placing hosts and

services into groups. Nagios provides methods for creating users and user groups. The notification section is highly configurable to meet your reporting needs. Nagios can monitor your environment in several ways.

1. Active Checks: Nagios will use protocols such as PING and HTTP to externally connect to devices. Nagios can also use SSH or NRPE to connect to a device and run a plugin (script).

2. Passive Checks: An outside process contacts Nagios.

Nagios comes with many built-in plugins and commands that will enable you to start monitoring servers very quickly.

Sample Nagios Configuration

Once you download and install Nagios on your central monitoring server, you will need to set up some configuration files. First define contacts and contact groups in the

Nagios contacts.cfg and contactsgroup.cfg file respectively. This will allow you to notify certain individuals or groups of people. You can set time ranges for notification as well as what types of severity alerts should be sent to which parties.

In this example, we will set up Nagios to ―ping‖ our Avamar nodes to determine if they are up. For simplicity you can add all your nodes into a file called avamar_hosts.cfg. This file defines ALL Avamar nodes.

 A typical entry looks like this for each node:

You can also create an avamar_group.cfg file that would group certain Avamar nodes together. For example, hostgroup ―avamar -servers‖ would contain ALL nodes, whereas hostgroup ―avamar -utility-servers‖ would only contain the utility nodes. A typical entry would look like this:

Now we can create a service configuration file that defines what we want Nagios to check for. This file contains default settings such as name of the service, what checks are allowed, if notifications are enabled, time period that checks are allowed, the contact group, polling interval, and so on.

In this example, we created a file called avamar_service.cfg. This file contains our

default settings for whenever this service is used. It also contains our ―ping‖ check. Here is an example of that check.

 As you can see, we are using the avamar-service default settings, we are applying this check against the hostgroup ―avamar -servers‖, the service description name will show up as PING in the browser, and last, we are calling the built-in command ―chec k-host-alive‖.

Once all configuration changes are made, the Nagios service needs to be reloaded using the /etc/init.d/Nagios reload command. If errors are detected, you can

troubleshoot by calling the Nagios binary and pointing to the configuration file, i.e. /usr/local/Nagios/bin/Nagios –v../etc/Nagios.cfg

Here is a screen shot of the PING service for our node ―localhost‖:

You can also drill down into the service for more detail:

Advanced Nagios Configuration

PLUGINS

Nagios can be set up to do very elaborate checks via plugins on the nodes. You might want to have Nagios check node capacity or GSAN status. The plugin on the node can return information to be displayed in the browser window and depending on the exit code, it can trigger the appropriate alert and notification. There are two ways that Nagios can connect to the server and run the plugin.

1. NRPE – this is a daemon process that is installed and runs on each server to be monitored

2. SSH – With the use of SSH keys, Nagios can connect to the server and run commands

In this example, we will connect to a Utility node using SSH and run a plugin to see if any of the nodes are above 60% utilization. On the Avamar Utility node, we have

created a script called ―nodeutil‖. The purpose of this script is to check each node and put Nagios into a warning state if the node utilization is between 60% and 62%. It will put Nagios into a critical state if the node utilization is 63% or higher. Here is what Nagios looks like after the check is run:

We can also connect to the utility node and run other plugins to check Avamar-specific services, for instance, the status of the GSAN. The logic of your plugin needs to be aware that the GSAN status changes based on the maintenance job that is running and should account for that. This prevents false Nagios alerts. Here is a screen shot of more checks that you can monitor:

PASSIVE CHECKS

Passive checks can also be used to update Nagios about events. In this example, we will tie together the SYSLOG/SWACTH setup used earlier to contact Nagios. Recall that SWACTH can run a script if a pattern is matched. Nagios has a built-in command that you can call and assign variables. In this case we have SWACTH call a homegrown script called ―alert_nagios‖ that looks at the error message to determine the node name and then calls the Nagios command ―submit_check_result‖.

Here is our SWATCH configuration to check for Garbage Collect failure:

Here is what Nagios looks like after the check runs:

You can also update Nagios the same way to show that a check was successful. Here is our SWATCH configuration to check for Garbage Collect success:

Here is what Nagios looks like after the check:

Conclusion

Monitoring Avamar includes not only the application itself but the devices that make up the Avamar infrastructure. Syslog and Nagios are free tools that provide limitless

options for building a solid central monitoring solution. These tools extend the packaged  Avamar features to create a comprehensive monitoring system. The added capabilities

and automated coverage help alleviate the time you spend worrying about backups.

Demystifying Avamar 

_—

Key Concepts and Features

 Arunkamar Velumani & Anay Pathak

[email protected], [email protected] Computer Sciences Corporation India Pvt Ltd

Disclaimer: The views, processes, or methodologies published in this article are those of the authors. They do not necessarily reflect EMC Corporation’s views, processes, or methodologies.

EMC believes the information in this publication is accurate as of its publication date. The information is subject to change without notice.

THE INFORMATION IN THIS PUBLICATION IS PROVIDED ―AS IS.‖ EMC

CORPORATION MAKES NO RESPRESENTATIONS OR WARRANTIES OF ANY KIND WITH RESPECT TO THE INFORMATION IN THIS PUBLICATION, AND SPECIFICALLY DISCLAIMS IMPLIED WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.

Use, copying, and distribution of any EMC software described in this publication requires an applicable software license.

Executive Summary

Data deduplication is a method for eliminating redundant data from storage, especially from backups. It works by saving a single copy of identical data, replacing any further instances with pointers back to that one copy.

Figure 1: Data Deduplication

When it comes to data deduplication, one size does not fit all. That's why it is important to consider a solution’s approach from the following three levels before making a

decision:

1. Where does data deduplication occur? Does it occur at the source (a server, for example) or at the target that stores the data (a virtual tape library, for example)?  A source-based approach results in less data being sent across the wire for

backup, potentially shortening backup windows.

2. When does deduplication happen? In target-based implementations, data can either be backed up first, then deduplicated (post-process), or Deduplication can be executed during the backup process (inline). Each method has pros and cons: Post-process Deduplication may result in a faster backup, but inline can be

replicated immediately after a backup concludes.

3. How does it happen? Object-level differencing reduces data by storing only the changes that occur, while hash-based chunking products locate global

redundancies that occur among all files in a backup. Some technologies even difference at the file level, a technique with so many drawbacks there's little point in considering it here.

With EMC Avamar ®, we get the advantage of source-based deduplication as well as target- based deduplication (with Data Domain® integration). It is an ideal solution for Enterprise backups as well as remote office backups (considering its great dedupe capabilities).

Abstract

Data Protection is major challenge for all organizations. Data is growing exponentially and robust backup technologies are required to meet Recovery Time Objective

(RTO)/Recovery Point Objective (RPO) associated with Backup and Recovery of customer data.

Moving from tape to disk is not a recommended solution as most companies experience some type of pain associated with tape; i.e.inability to restore in an acceptable time frame; physical nature of tape; snapped tape; unreadable tape; slow Serial Access devices; and so on.

With drastic data growth, adding disk is not a good solution considering the cost

associated with it. Thanks to deduplication, a major revolution is taking place in the data protection arena. Gartner views this technology as transformational because it radically decreases the economics of disk-based backup and recovery.

There are a number of deduplication vendors and it is always good to look at deduplication ratios when comparing different products which are available. Deduplication should save you much more space than compression.

So, what options exist to help your IT department with this challenge?

Of the many products in market, a key player is EMC  Avamar ® which has transformed the backup technology over the last few years. Taking advantage of data deduplication, this software/hardware tapeless backup and recovery solution deduplicates the data at the source client on a global level. Avamar  offers:

 Global data deduplication  Systematic fault tolerance

 Highly reliable, inexpensive disk storage  Standard IP network technologies

 Scalable server architecture  Flexible deployment

 Centralized management  Desktop/Laptop support  VMware protection

 Efficient monitoring and reporting  Tape Out / Import

 Fastest recovery

 Remote Office backup

 Less network bandwidth consumption

This article intends to brief challenges faced by organizations backing up and recovering data along with Avamar  key concepts and features,

including Avamar  integration with Data Domain and NetWorker ®, VMware advantages, NDMP enhancements, Home Base integration, remote office data protection, and so on. Those who read this article will gain a comfort level about key concepts and

instrumental features of Avamar  and will be able to use them efficiently to solve their data protection challenges.

Abstract

Data Protection is major challenge for all organizations. Data is growing exponentially and robust backup technologies are required to meet Recovery Time Objective

(RTO)/Recovery Point Objective (RPO) associated with Backup and Recovery of customer data.

Moving from tape to disk is not a recommended solution as most companies experience some type of pain associated with tape; i.e.inability to restore in an acceptable time frame; physical nature of tape; snapped tape; unreadable tape; slow Serial Access devices; and so on.

With drastic data growth, adding disk is not a good solution considering the cost

associated with it. Thanks to deduplication, a major revolution is taking place in the data protection arena. Gartner views this technology as transformational because it radically decreases the economics of disk-based backup and recovery.

There are a number of deduplication vendors and it is always good to look at deduplication ratios when comparing different products which are available. Deduplication should save you much more space than compression.

So, what options exist to help your IT department with this challenge?

Of the many products in market, a key player is EMC  Avamar ® which has transformed the backup technology over the last few years. Taking advantage of data deduplication, this software/hardware tapeless backup and recovery solution deduplicates the data at the source client on a global level. Avamar  offers:

 Global data deduplication  Systematic fault tolerance

 Highly reliable, inexpensive disk storage  Standard IP network technologies

 Scalable server architecture  Flexible deployment

 Centralized management  Desktop/Laptop support

 VMware protection

 Efficient monitoring and reporting  Tape Out / Import

 Fastest recovery

 Homebase backup & recovery  Remote Office backup

 Less network bandwidth consumption

This article intends to brief challenges faced by organizations backing up and recovering data along with Avamar  key concepts and features,

including Avamar  integration with Data Domain and NetWorker ®, VMware advantages, NDMP enhancements, Home Base integration, remote office data protection, and so on.

Those who read this article will gain a comfort level about key concepts and

instrumental features of Avamar  and will be able to use them efficiently to solve their data protection challenges.

Avamar  Terminologies and System Components

Avamar  server / Grid

The Avamar  server is a group of one or more Nodes on a local, high-speed network. It stores client backups and provides essential processes and services required for client access and remote system administration. Avamar  Administrator Server (mcs)

and Avamar  Data Server (gsan) run on the Avamar  server.

The three Avamar  server editions provide the flexibility to meet different customer requirements.

 Software Only

The server software is installed on customer-supplied Avamar -qualified hardware platforms. Currently, Dell, HP, and IBM hardware platforms are supported.

 Hardware Appliance ( Avamar  Data Store)

Consists of Avamar server software installed onsite on configured and pre-tested Avamar -certified hardware.

 Virtual Appliance ( Avamar  Virtual Edition)

 Avamar  server will be installed on VMware ESX server Virtual Machine. It will be single node non-RAIN server up to maximum capacity of 2TB.

Avamar  client

 A System is one or more Avamar  servers and the network servers or desktop clients that back up data to those servers. Avamar  Client software runs on each computer or network server being backed up. Avamar  provides client software for various computing platforms.

Avamar  A d m i n i s t r a to r

 Avamar  Administrator is a user management console software application that is used to remotely administer an Avamar  system from a supported Windows or client

computer.

Global Storage Area Networ k

The Global Storage Area Network (GSAN) is a component of the  Avamar  server. It is sometimes referred to as the data server or the storage server.

Node

 A Node is a self-contained, rack-mountable network-addressable computer consisting of both processing power and hard drive storage. Nodes run Avamar  server software on the Linux operating system.

Utility node

Utility node is dedicated to providing internal Avamar  server processes and services, including the administrator server, cron jobs, Domain Name Server (DNS), external authentication, Network Time Protocol (NTP), and web access.

Data storage node

Data storage nodes include the Avamar  Data Server software and are dedicated to providing backup storage.

NDMP accelerator no de

The NDMP accelerator is a specialized node that, when used as part of

an Avamar  system, provides a complete backup and recovery solution for NAS devices via the Network Data Management Protocol (NDMP).

Media access node

The Avamar  extended retention feature, included in version 6.1 of the

core Avamar  software product, is hosted on a dedicated media access node which is used for Avamar  tape import and export operations and has a different part number than ADS. Gen4 7.8TB node has usable capacity of 6.5 TB of deduplicated storage. A single media access node supports up to five Servers.

Transport node

Transport nodes are Avamar  Virtual Edition servers that have been modified specifically for use in the Avamar  Data Transport system. They are the targets of replication

from Avamar  servers, and are used as backup and restore targets by the tape backup server. Transport nodes are sized at 1TB or 2TB nodes.

Control node

The control node is a 64-bit Linux host where the Avamar  Data Transport application and the Avamar  Data Transport Framework run. The Avamar  Data Transport user interface is accessed from a web server that is installed on this host. In addition, the database of all transported files is often located on the control node. There is only one control node in an Avamar  Data Transport system.

Tape backup server / Arch ive server

 Avamar  Data Transport supports the EMC NetWorker or Symantec NetBackup tape backup products. Both products have a server component that drives the tape archiving process.

Import server

Import server is the Avamar  server that is installed on the Media Access node. Backups are imported from tape to this server.

Avamar  Data Transp ort

 Avamar  Data Transport is a system of virtual machines that run in a VMware ESX Server environment in conjunction with tape backup software to transport

deduplicated Avamar  data to tape.

Avamar  Extended Retention

The Avamar  extended retention feature is a solution that exports backup data from an Avamar  server to a tape library (physical or virtual) for long-term retention purposes. This backup data can be imported (or restored) from tape as needed.

A g e n t s

 Avamar  agents are platform-specific software processes that run on the client and communicate with the Management Console Server (MCS) and any plug-ins installed on that client.

There are two basic types of Avamar  plug-ins:

 Filesystem plug-ins are used to browse, back up, and restore files or directories

on a specific client file system

  Application plug-ins support backup and restore of databases or other special

applications

Enterprise Manager

The Avamar  Enterprise Manager provides centralized access to

the Avamar  Administrator for each Avamar  system in an enterprise, as well as dashboard, reporting, and search capabilities.

Client Manager

Client Manager is a graphical interface used to activate, manage, and analyze backup clients. It is especially useful when dealing with an environment with a large number of clients to back up and manage.

AvInstaller progr am

The AvInstaller program is included in the Avamar  software bundle. For new

installations of Avamar  v6.x, AvInstaller software and the Avamar  Installation Manager User interface are installed on the single node Avamar  server.

EMC repository

EMC repository contains server, client installation packages, and manifest files. The EMC repository currently spans two locations: ftp://ftp.avamar .com and EMC

Subscribenet. High urgency packages such as hot fixes and OS patches are available from ftp://ftp.avamar .comftp://ftp.avamar .com

Low urgency packages are available from EMC Subscribenet.

Local repository

Local repository is the /data01/avamar /repo/packages directory on the Avamar  single node server. The directory contains the most current manifest file from the EMC

repository, installation packages, and workflow packages. The Avamar  Downloader Service pushes packages from the EMC repository to the local repository. If a customer site does not allow Internet access, packages can be manually copied into the local repository.

Manifest file

 An XML file listing all packages currently available for download from the EMC

repository. When the EMC Subscribenet team adds a package to the EMC repository, it then adds an entry to the manifest file that describes the package and vice versa.

Avamar  Inst allation Manager

 Avamar  Installation Manager is a web interface that manages installation packages. A successful Avamar  v6.1 server software installation or upgrade embeds

the Avamar  Installation Manager Functionality in the Avamar  Enterprise Manager. This feature is called System Maintenance.

Avamar  Dow nloader Service

 A Windows-based file distribution system that delivers software installation packages to target Avamar  systems. The Avamar  Downloader Service automatically downloads the manifest file from the EMC repository once a day and determines if new download packages are available.

Node resou rce database

 An XML file named probe.xml stores the types and IP addresses of Avamar  server nodes. It also stores the types and IP addresses of optional nodes such as media access node.

Stripe

 A Stripe is a unit of disk drive space managed by Avamar .

Object

 An Object is a single instance of deduplicated data. Objects are stored and managed within stripes on the Avamar  server.

Domains

Domains are distinct zones within Avamar  that are used to organize and manage backup clients. They are used to manage administration access to groups of clients

G r o u p s

Groups are used to create automated backups for a set of clients. Avamar  backup runs according to group specifications.

E x p o r t p o l i c y

Defines the schedule on which a group of client backups will be exported to tape

by Avamar  extended retention. The export policy defines the group of clients that share a common schedule and retention periods for backup metadata and file-level metadata.

Datasets

Datasets define the persistent backup selections for the file systems, directories, or files to be included in a backup

Schedules

The Schedule for a group determines when and how often a backup will automatically be run.

Retention Policies

Retention Policies specify how long each backup from the group will be kept.

B a c k u p w i n d o w

The backup window is when the majority of backups are performed. Backups should be scheduled to run during this time.

B l a c k o u t w i n d o w

During the blackout window, backups cannot be performed. This window of time is

reserved for maintenance activities such as garbage collection and checkpoint creation. By default, the blackout window runs from 8 am to 11 am.

Maintenance window

The maintenance window is reserved for other maintenance activities, primarily the HFS check. Backups may be initiated, but both backup time and maintenance activities will be impacted. By default, the maintenance window runs from 11am to 8 pm.

Avamar  Functions

Data Deduplication

Data deduplication is a key feature of the Avamar  system. Data deduplication ensures that each unique sub-file, variable length object is stored only once across sites and servers. This feature significantly reduces network traffic and provides greatly enhanced storage efficiency on the server.

B a c k u p

 An Avamar  backup is defined as a point-in-time copy of client data that can be restored as individual files, selected directories, or entire file systems. Avamar  provides two backup types: scheduled and on-demand.

 Scheduled backups are run automatically according to specifications that can be

 On-demand backups can be initiated from the Avamar  Administrator interface

and the Management Console command line interface (MCCLI). On-demand backups can also be run from a client.

Restore

Restore is an operation that retrieves one or more file systems, directories, or files from an existing backup and writes it to a designated location on clients.

E n c r y p t i o n

Encryption provides enhanced security during client/server data transfers and on the Avamar  server. An Avamar  server can be configured to encrypt all backup data stored on the server ―at rest‖.

 Avamar  can also encrypt data sent between client and server ―in-flight‖. Two types of in -flight encryption are available—medium and high. While it can also be disabled entirely, ―at rest‖ encryption can be enabled when Avamar  servers installed.

Replication

Replication is the process of transferring data from source to a

destination Avamar  server to support future disaster recovery of the source

server. Avamar  uses sophisticated data deduplication technology which eliminates data duplication and sends only unique data to the destination Avamar  server. Avamar  offers two types of replications.

 Normal replication

Replication automatically creates a REPLICATE domain on destination server during first replication which is a mirrored representation of the entire source

server client tree. During this normal replication, user data from source Avamar  is replicated to destination Avamar  server.

 Full root-to-root replication

This replication creates a complete logical copy of an entire source server on the destination server. This data is not copied to REPLICATE domain on destination server; it is added directly to root domain as if the source clients had registered with the destination server.

Checkpoint

 A checkpoint is a read-only snapshot of the Avamar  server taken to enable server rollbacks. Checkpoints are created using hard-links to all the stripes.

Hash File System c heck

 A hash file system (HFS) check is an operation that validates the integrity of a

checkpoint. Once a checkpoint has passed a HFS check, it can be considered reliable enough to be used for a system rollback. Checkpoints are taken twice daily and

Garbage collection

Garbage collection is the process of deleting unused chunks from backups that have expired. Garbage collection requires a quiet system in order to run. While garbage

collection is running, the system is placed in a read-only condition. Garbage collection is run once a day during the blackout window.

Section II: Avamar  Data Protection Enhancements

Microsoft Exchange Integration

Microsoft Exchange Server is the server side of a client –server, collaborative application product developed by Microsoft. Exchange's major features—electronic mail,

calendaring, contacts, and tasks—work with Microsoft Outlook on PC and Mac.  A Database Availability Group (DAG) is a set of up to 16 Microsoft Exchange Server

2010 Mailbox servers that provides automatic, database-level recovery. DAGs use continuous replication and a subset of Windows failover clustering technologies to provide high availability and site resilience

 A Recovery Database (RDB) is used for recovering Exchange 2010 mailbox database data and specific email items. RDB has to recover data from a backup without

disturbing user access to the current data.

Microsoft Exchange 2010 Backup Implementation

Below are the steps for Exchange 2010 VSS Backup Configuration

1. Install .NET4.0 and MAPI (Microsoft Messaging API) / CDO (Collaboration Data Objects) on all DAG nodes.

2. Install Avamar  file system and Exchange VSS plug-in on all DAG nodes. Optionally, you can install GLR feature which needs a reboot.

Note Please make sure to mount the active mailbox databases to alternate_{node prior installing and rebooting the  Avamar  file system, Exchange}

VSS, and GLR.

3. Run AvamarBackupUser Configuration tool on one of the Exchange servers and create an AvamarBackupUser account.

4. Change Avamar  Services on all DAG nodes to run under AvamarBackupUser account and restart Backup Agent service.

5. Register all the Clients on DAG.

6. Run the Avamar  Cluster configuration tool from one of the Exchange servers for installing and configuring the Avamar  Exchange DAG clients and provide static IP address and var directory for Exchange DAG client.

7. Run an on-demand full or incremental backup from Avamar  Administrator

Backup & Restore GUI by selecting the DAG client and Windows Exchange VSS plug-in. Backups can be automated like any other file system backup by selecting the DAG name as backup client. Other backup options such as active or passive replica writers, server preferred order list, and circular logging can also be

Note Exchange Server 2010 supports only Microsoft Volume Shadow Copy_{Service (VSS)-based backup and Extensive Storage Engine (ESE)}

streaming backup has been discontinued from Exchange Server 2010.

Microsoft SharePoint Integration

Microsoft SharePoint is a Web application platform developed by Microsoft. The

common solutions using SharePoint are collaboration, content management, enterprise search, and web portals.

 A SharePoint server is an individual server that runs the operating system and

application software required to perform roles or to provide services for the SharePoint farm.

 A SharePoint farm is a set of servers comprised of databases, applications, and web services that together provide a SharePoint solution.

 A Web-Front-End (WFE) server provides services directly to end users and handles all web traffic from end users.

Microsoft SQL server of the content and settings in a SharePoint server are stored in Microsoft SQL relational databases.

Microsoft SharePoint 2010 Backup Implementation

Microsoft SharePoint Server 2010 provides two backup systems: farm and granular. Components in the farm that can be selected for backup include the farm, web

applications, and shared and non-shared services and service applications. With the granular backup system, a user can back up a site collection or export a site or list.  Avamar  provides two different plug-in for SharePoint

 Windows SharePoint Plug-in - One web front end server must be selected as

owner of the backup and all other backend / SQL servers must be installed in secondary mode.

 Windows SharePoint Granular Recovery (GLR) Plug-in - Used for granular level

recovery.

Microsoft SharePoint Backup Configuration Steps in Avamar 

1. Install Avamar  Windows file system client. For GLR configuration, set the Backup  Agent service to run on SharePoint administrator credentials.

2. Install Avamar  SharePoint VSS plug-in and select install GLR feature on hard disk for GLR configuration. On any one of the web front end servers, choose front-end plug-in. Other web front end servers do not have plug-ins; every other server in the farm is installed as back-end plug-in.

3. Activate all clients with Avamar .

4. From the Avamar  Backup & restore GUI, select the server that has the front end plug-in installed and choose the Windows SharePoint VSS plug-in. Select Farm, right click, and select Backup up now to back up the SharePoint Farm.

Microsoft SQL Integration

Microsoft SQL Server is a relational database management system developed by

Microsoft. As a database, it is a software product whose primary function is to store and retrieve data as requested by other software applications. Avamar  supports full,

differential, and incremental backup (only for transaction log backups). For SQL, it supports multi streaming up to six.

Microsoft SQL Backup Implementation for Windows servers

1. Install the Avamar  for Windows client and SQL in for SQL clients (SQL plug-in needs Microsoft .NET 4 plug-installed) and register with Avamar  Server.

2. From the Avamar  Backup & restore GUI, select the SQL server and choose the Windows SQL plug-in. Select the database then right click and select Backup up now to back up the database.

3. When you select More Options, you can choose the backup type, force full backup (runs full backup if no previous full backup is available), use Data Domain® system (if any Data Domain is configured), and so on.

NDMP Integration

Network Data Management Protocol (NDMP) is an open protocol used to control data backup and recovery communications between primary and secondary storage in a heterogeneous network environment.

Figure 3: Avamar  environment with NAS backup using Accelerator node

NDMP Backup Implementation steps for NetApp

 Avamar  uses NDMP Accelerator node to backup NDMP storage devices. Accelerator handles the data deduplication and NDMP processing and forwards the data

to Avamar  Server.

Steps for Configuring Backu ps for NetApp

1. Install Avamar  server, accelerator, and filer on the network and ensure all three have connectivity each other.

2. Log in to filer and run options ndmpd.access host=HOSTNAME to provide access to the accelerator and run options ndmpd.enableto enable NDMP services on the filerwhere HOSTNAME is accelerator hostname.

3. Download the Avamar  client for Linux and accelerator from Avamar  Server and install them on the accelerator as root.

4. Log in as root on accelerator. Run avsetupndmpand follow the instructions to complete configuring the accelerator. Be sure to enter the NDMP user account when prompted, which has sufficient privileges to run the NDMP jobs.

5. Run the avregisteronaccelerator to register it to the Avamar  Server.

6. In Avamar  Administrator, click the Backup & Restore launcher button. The Backup and Restore window appears.

7. In the client’s tree, select a storage device client then select for Backup tab. 8. Expand the accelerator node in the tree in the left pane of the Select for Backup

tab.

9. Select the checkbox next to the volumes to back up.

10. Select Actions > Backup Now to back up selected volumes.

Oracle Integration

 An Oracle server consists of an Oracle instance and an Oracle database. The Oracle instance is software executed on the server as part of the Oracle server that stays resident in memory. The database consists of files that are located on any number of physical storage devices. These files are timestamp-related to each other; losing one file results in a corrupt database. A database server is the key to solving the problems of information management.

Oracle Implementation with Avamar 

The Avamar  Plug-in for Oracle works with Oracle and Oracle Recovery Manager

(RMAN) to back up an Oracle database, a tablespace, or datafiles to an Avamar  server. Backup and recovery operations are initiated from RMAN.

Steps to configure RMAN backups for Linux servers:

1. Install Avamar  Client for Linux and Oracle plug-in on the Oracle server.

2. To back up Oracle RAC databases, run the rac_config script on all cluster nodes, register only one node with Avamar  Server, and select shared location for var directory.

3. Create backupuser on an Oracle database and provide sysdba

privilege. Avamar  server uses this account to perform backup and restores.

4. Performing hot backups database, youe must be in ARCHIEVELOG mode. To do so, run ―alter database archivelog;‖ after connecting to the database.

5. In Avamar  administrator, open the Backup & Restore window and select the Oracle server.

6. Select the database to back up, right click and select Backup up now to back up the databases selected.

NetWorker Integration

EMC NetWorker ® is an Enterprise backup and recovery software that centralizes, automates, and accelerates data backup and recovery in an IT environment.

NetWorker Integration with Avamar 

1. Install NetWorker Linux Client software on Avamar  utility node and start NetWorker daemons.

2. Stop the MCS, edit the /usr/local/avamar /var/mc/server_data/prefs/mcserver.xml file, and set the allow_duplicate_client_names attribute to true.

3. Restart MCS and log in to Avamar  Administrator to create a domain called NetWorker.

4. Go to the NetWorker Administration Devices window, right-click Deduplication Nodes, and select New. In the Identity area, enter the name of

the Avamar  server. In the Access area, enter the user name and password for the Avamar  server.

5. A NetWorker advanced file type device (AFTD) is required for storing deduplication metadata. To create an AFTD device; from the NetWorker

 Administration Devices window, right-click Devices and, from the pop-up menu, select New. In the Name area, enter the path for the device. Then, select adv_file for Media type.

6. Label the deduplication AFTD in the deduplication pool and mount the device. 7. Create a dedicated group for deduplication clients and tie the deduplication pool

to the group.

8. Go to the NetWorker Administration Configuration window, right-click Clients and click New. Provide the server name then select the Deduplication group

and Avamar deduplication node to be used for backups. Make sure to check the Avamar  deduplication backup checkbox.

9. Now you can run the Avamar  deduplication backup just as you would run traditional NetWorker backups.

Data Domain Integration

MC Data Domain is a target-based deduplication system that uses inline deduplication whereas Avamar  uses variable segment deduplication for data deduplication. A Data Domain system manages data of greater than 5TB more effectively than Avamar  server, especially for large, active databases.

 Avamar  clients use Data Domain Boost API (Application Programming Interface) to access and store data to a data domain system. This API is installed automatically when Avamar  client is installed on the client computers. Avamar  plug-ins that support backup and restore to and from Data Domain include: IBM DB2, Microsoft Exchange VSS, Hyper-V VSS, SharePoint VSS and SQL Server, Oracle, SAP with Oracle, Sybase, and VMware image backup and restore.

Figure 4: Avamar  environment with Data Domain DD Boost backup

Data Domain Implementation steps with Avamar 

1. Log in to Data Domain CLI (Command Line Interface) as administrative user and enable DD boost by typing ddboost enable

2. Create a DD boost user account and set it as DD boost user by typing

 user add user priv admin  ddboost set user-name user

where user is username for the account

3. Disable and Enable the DD Boost for the changes to take effect by typing

 ddboost disable  ddboost enable

4. Enable SNMP if not already enabled from Data Domain Enterprise Manager  System Settings  SNMP

5. Log in to Avamar  server GUI with administrative privileges and go

to Avamar  Administrator Server  Server management tab  Actions  Add Data Domain Systems, then supply the fully qualified domain name of the Data Domain system and the DD Boost user name and password. On the SNMP tab, supply the SNMP Community String used to grant Avamar  read and write access to the Data Domain system, then click OK.

6. In Avamar  administrator, open the Backup & Restore window and select the server that has one of the Data Domain supported plug-ins installed (for example, Microsoft SQL) then select the plug-in and right click Back Up

Now More Options. On the Backup Command Line Options dialog box, select Data Domain system that has been added to the Avamar  server, then click OK to run on-demand backups to the selected Data Domain system. For a scheduled backup, this is done from the Options tab in a Dataset specification.

Note DD Boost is a licensable feature in Data Domain and DD Boost_{becomes the preferred method of connectivity for any clients that are}

enabled for DD Boost while DD Boost is also enabled on the Data Domain.

DB2 Integration

IBM DB2 is a relational model database server developed by IBM. There are three main products in the DB2 family: DB2 for LUW (Linux, Unix, and Windows), DB2 for z/OS (mainframe), and DB2 for iSeries (formerly OS/400).

DB2 Integration with Avamar 

The Avamar  Plug-in for DB2 supports backups and restores of DB2 single instance configurations and DB2 configurations that implement DB2 Database Partitioning

Feature (DPF). Each DB2 server requires the installation of the Avamar  Plug-in for DB2 and an Avamar  file system client. You can back up and restore DB2 databases

using Avamar  Administrator.

Configuration steps for DB2 backups on AIX servers

1. Log in to the DB2 server host as root and download AIX and DB2 installation package.

2. Install and register AIX client package to Avamar  server. Once done, we need to install and configure DB2 as given below:

a) Install Avamar  Plug-in for DB2

geninstall -d AvamarDB2-aix5-ppc64-VERSION.bff all

b) Configure the Avamar  Plug-in for DB2 software by typing: /usr/local/avamar /bin/avdb2setup

c) Enter the Username for DB2 instance owner when prompted for the same. d) Provide Home directory for DB2 instance owner.

3. To perform first online backup of DB2 database:

 For DB2 version 9.1 and earlier, type: db2 update db cfg for

DATABASE-NAME using LOGRETAIN ON where DATABASE-DATABASE-NAME is the name of the database to back up.

 For DB2 version 9.5 and later, type: update db cfg for

DATABASE-NAME using LOGARCHMETH1 LOGRETAIN where DATABASE-DATABASE-NAME is the name of the database to back up.

Sybase Integration

The Avamar  Plug-in for Sybase plug-in is software that works with the  Avamar  server and client software to provide deduplication backup and recovery for Sybase Adaptive Server Enterprise (ASE) data. A Sybase ASE server is the only type of Sybase

database server that the Sybase plug-in supports.

The Sybase plug-in is deployed in an Avamar  client/server system in a network

environment. The Sybase server host is Avamar  clients that access the Avamar  server for backup and restore services.

Configuration steps for Sybase backups with Avamar  on Linux

1. Download, install, and register client package as root to Avamar  server as given below:

e) Type the following command depending on your Hardware: rpm -hi  AvamarSybase-linux-type-x86-version.rpm

f) Create a symbolic link named libsybase_avamar.so  in the Sybase ASE library directory, $SYBASE/$SYBASE_ASE/lib, that points to the Avamar  Plug-in for Sybase library file, libsybase_avamar.so, in the Avamar_installation_dir/lib directory.

g) Edit the avsybase script file, and add the Sybase OCS (Open Client Server) library path to the LD_LIBRARY_PATH  path in the script.

h) Register the Sybase plug-in by running /usr/local/ avamar  /bin/avregister  2. In Avamar  Administrator, click the Backup & Restore launch button and select

the Sybase server that needs to be backed up.

3. Select the Sybase plug-in type and complete the settings in the Browse Command Line Options dialog box:

 In the Sybase installation directory field, type the full pathname of the

Sybase installation directory for the server to be backed up, represented by $SYBASE on Linux.

Note type the actual text of the full pathname. For example, /sybase_{instead of $SYBASE.}

 In the Sybase server name field, type the Sybase server name, not the

hostname. For example, type SERVER_ABC.

 In the OCS library directory field, type the full pathname of the Sybase

OCS library directory, represented by $SYBASE/$SYBASE_OCS/lib on Linux or UNIX.

Note Type the actual text of the full pathname. For example, type_{sybase/OCS-15_0/lib instead of $SYBASE/$SYBASE_OCS/lib.}

 In the Sybase username and password field, type the Sybase username

and password, then click OK.

4. Select Backup tab which displays a folder with the Sybase server name and perform one of the following on the Select for Backup tab:

 To back up all the databases on the server or the logs of all the

databases, select the checkbox next to the Sybase server name.

 To back up specific databases or the logs of specific databases:

a. Double-click the Sybase server name.

 All the databases of the Sybase server appear in the right pane.

b. Select the checkboxes next to the databases that you want to include in the backup.

5. Select Actions Backup Now and then give values for Retention and encryption as per standard configuration in Backup Environment. Click More Options and configure below parameters:

a) Backup Command Line Options dialog box appears. Complete the settings in the Backup Command Line Options.

b) To view the advanced options with red labels, select Show Advanced Options. c) Complete the settings in the Sybase Information group box:

 – In the Sybase installation directory field, type the full pathname of the Sybase installation directory for the server to be backed up, represented by $SYBASE on Linux or UNIX

 – In the OCS library directory field, type the full pathname of the Sybase OCS library directory, represented by $SYBASE/$SYBASE_OCS/lib on Linux or UNIX d) In the Sybase username field, type the Sybase username. In the Sybase user password field, type the Sybase user password.

e) If required, select Enable debugging messages and select Store backup on Data Domain system to store the backup data on a Data Domain system. f) Select the backup type as: Full backup, Incremental backup, Incremental backup with -no_truncate option, then Log truncation with truncate_only option followed by full backup; Log truncation with no_log option followed by full backup and next, complete the General settings for Number of Streams, Backup data verification -No verification, Full verification, Header verification and, if required, Protected backup password field. Type the password that the Sybase server uses with the Sybase dump and Pre and Postprocessing Options.

6. Click OK to close the Backup Command Line Options and Click OK to close the On Demand Backup Options dialog box and start the backup. Backups can be scheduled to run by creating Dataset for Sybase backups.

VMware Integration

 Avamar  for VMware image backup and restore is built on the VMware vStorage API for Data Protection (VADP). Avamar  for VMware image backup and restore is fully

integrated with VMware vCenter Server to provide easy detection of virtual machine clients within the vCenter, and enable efficient centralized management of backup jobs.

 Avamar  utilizes Changed block tracking as well which is a VMware feature that tracks which specific file system blocks on a virtual machine have changed between backups. Using this feature provides valuable I/O reduction earlier in the backup process.

Backups and restores require the use of proxy virtual machine clients. Each proxy virtual Machine client provides all of the following capabilities:

*Backup of Microsoft Windows and Linux virtual machines (entire images or specific drives)

*Restore of Microsoft Windows and Linux virtual machines (entire images or specific drives)

*Restore of individual folders and files to Microsoft Windows and Linux virtual machines.

Proxies run Avamar  software inside a Linux virtual machine, and are deployed using an  Appliance template (.ova) file.

Figure 5: VMware backup environment using Avamar 

Steps for VMware Integration with Avamar  (Windows)

 Avamar  for VMware image backup and restore supports protecting up to 5 vCenters from a Single Avamar  server. Beginning with Avamar  6.0, support for multiple vCenters is enabled by default during new Avamar  server software installations. However, if your Avamar  Server was upgraded from the previous version, you might need to perform the following manual configuration to enable support for multiple vCenters.

To enable support for multiple vCenters

1. Open a command shell and log in using one of the following methods: * To log in to a single-node server, log in to the server as admin.

* To log in to a multi-node server, log in to the utility node as admin.

2. Change directories by typing: cd /usr/local/ avamar  /var/mc/server_data/prefs 3. Edit com.avamar .mc.VMware.max_number_of_vcenters  in mcserver.xml as

shown below: <root type="system"> <node name="com">

<node name="avamar "> <node name="mc">

<node name="VMware">

<entry key="max_number_of_vcenters" value="1" />

Note Substantial portions of mcserver.xml have been omitted for clarity.

4. Change the max_number_of_vcenters  entry to an integer value between 2 and 5.

Note This value must be equal to or greater than the maximum number of_{vCenters you will be using with this  Avamar  server. Currently, the}

absolute maximum number of vCenters that can be supported by a single MCS is 5.

5. Save Changes and restart MCS by typing: dpnctl stop mcs

dpnctl start mcs

6. Download and install vSphere client software.

7. Download and install Avamar  Administrator software. 8. Configure vCenter-to- Avamar  authentication.

9.  Avamar  VMware Image Backup will not work unless:

* A valid authentication certificate is present on the Avamar  Management Console Server.

* Certificate authentication for all MCS-to-vCenter communications is turned off. Note: You must perform this task for each vCenter you intend to protect. 10. Open a command shell and log in

* If logging into a single-node server, log in to the server as admin. * If logging into a multi-node server, log in to the utility node as admin. 11. Stop the MCS by typing:

dpnctl stop mcs and switch to root su

-12. Copy rui.crt  from the vCenter machine to /tmp on the utility node or single-node server. The default certificate provided with vCenter is:

* Windows 2008: C:\ProgramData\VMware\VMware VirtualCenter\SSL\rui.crt * Other Windows versions: C:\Documents and Settings\All

Users\ApplicationData\VMware\VMware VirtualCenter\SSL\rui.crt

* Linux: /etc/VMware-vpx/ssl/rui.crt

13. Create a temporary version of the MCS keystore by copying the live keystore to /tmp and typing: cp /usr/local/ avamar  /lib/rmi_ssl_keystore /tmp/ 

14. Add the default vCenter certificate to the temporary MCS keystore file by typing: cd /tmp

-keystore rmi_ssl_keystore

Note: wher ALIAS is a user-defined name for this certificate, which can often be the file name.

15. When prompted for a password, type the root password. The following appears in the command shell:

Trust this certificate? Press yes and enter 

16. Back up the live MCS keystore by typing: cd /usr/local/ avamar  /lib

cp rmi_ssl_keystore rmi_ssl_keystore.DATE Note: where DATE is today’s date. 17. Copy the temporary MCS keystore to the live location by typing: cp

 /tmp/rmi_ssl_keystore /usr/local/ avamar  /lib/ 

18. Exit the root subshell, and restart the MCS by typing: exit 

dpnctl start mcs

19. Create a dedicated vCenter user account. EMC strongly recommends that you set up a separate vCenter user account that is strictly dedicated for use

with Avamar .

Note You must perform this task for each vCenter you intend to protect.

20. Add vCenter client in Avamar  Administrator by going to Actions > Account Management > New Client.

21. Backups and restores require deployment of proxy virtual machine clients. 22. Once deployed,each proxy virtual machine client provides all of the following

capabilities:

* Backup of Microsoft Windows and Linux virtual machines (entire images or specific drives)

* Restore of Microsoft Windows and Linux virtual machines (entire images or specific drives)

* Selective restore of individual folders and files to Microsoft Windows and Linux virtual machines

Proxies run Avamar  software inside a Linux virtual machine, and are deployed using an appliance template (.ova) file. Proxy clients are allowed in an y part of the Avamar  Administrator account management tree except the vCenter server domain or sub domains.

23. Once the Proxy is configured, in Avamar  Administrator, click the Backup & Restore launcher button. The Backup and Restore window appears. Click the Select for Backup tab and select a virtual machine client in the client’s tree. 24. In the Browse for File, Folders, or Directories pane, do one of the following:

* Select the top (root) folder to back up the entire image.

* Select one or more disks to only back up those specific virtual disks.

25. Click Backup Now to run the backup after reviewing your backup settings; i.e. Retention, encryption, and so forth.

Hyper-V Integration

Microsoft Hyper-V is a hypervisor-based server virtualization product for Microsoft Windows Server 2008 and Microsoft Windows Server 2008 R2 that enables you to create multiple virtual machines (VMs) on a single physical server to consolidate workloads.

With Hyper-V, the server is separated into individual partitions that are supported by the hypervisor. The parent partition performs all VM and hardware management. From the parent partition, you create a child partition for each VM (virtual machine). Possible

configurations for Avamar  Integration with Hyper-V: Hyper-V in a cluster, Granular Level Recovery, using Avamar  with Data Domain as target storage device.

You can perform both application and crash consistent image-level backups with the Avamar  Plug-in for Hyper-V VSS:

1. With an application consistent backup, the VM must be online and be VSS capable. Microsoft Integration Components (IC) must be installed. VSS runs in-guest and freezes the OS and all applications state.

2. With a crash consistent backup, the VM is offline or does not have the IC installed. In this case, the VM is paused before shadow copy creation and resumed after the shadow is created.

Possible backup workflows include:

Image-level VSS backup, Image-level saved state backup (or offline), Avamar  federated architecture for Hyper-V failover clusters, Image-level backup in Hyper-V federated

backups.

Steps for Hyper-V Integration with Avamar  on Windows servers

In this section we will see how to configure Hyper-V backups on Windows Servers 1. Log in to the Windows server host as administrator and download

the Avamar  client software for Windows. Install the software and register it to the Avamar  Grid.

2. Install the Avamar  Plug-in for Hyper-V VSS. To install the Avamar  Plug-in for Hyper-V VSS on a stand-alone Hyper-V Server, download AvamarHypervVSS-windows-x86_64-6.1.100.x.msi and install it on Hyper-V Management OS.